HISTORICAL REVIEW Biological Warfare at the 1346 Siege of Caffa Mark Wheelis*

On the basis of a 14th-century account by the Genoese Gabriele de’ Mussi, the Black Death is widely believed to have reached Europe from the Crimea as the result of a biological warfare attack. This is not only of great historical interest but also relevant to current efforts to evaluate the threat of military or terror- ist use of biological weapons. Based on published translations of the de’ Mussi manuscript, other 14th- century accounts of the Black Death, and secondary scholarly literature, I conclude that the claim that bio- logical warfare was used at Caffa is plausible and provides the best explanation of the entry of plague into the city. This theory is consistent with the technology of the times and with contemporary notions of dis- ease causation; however, the entry of plague into Europe from the Crimea likely occurred independent of this event.

he Black Death, which swept through Europe, the Near little help; they refer repeatedly to an eastern origin, but none T East, and North Africa in the mid-14th century, was prob- of the reports is firsthand. Historians generally agree that the ably the greatest public health disaster in recorded history and outbreak moved west out of the steppes north of the Black and one of the most dramatic examples ever of emerging or Caspian Seas, and its spread through Europe and the Middle reemerging disease. Europe lost an estimated one quarter to East is fairly well documented (Figure 1). However, despite one third of its population, and the mortality in North Africa more than a century of speculation about an ultimate origin and the Near East was comparable. China, India, and the rest further east, the requisite scholarship using Chinese and cen- of the Far East are commonly believed to have also been tral Asian sources has yet to be done. In any event, the Crimea severely affected, but little evidence supports that belief (1). clearly played a pivotal role as the proximal source from A principal source on the origin of the Black Death is a which the Mediterranean Basin was infected. memoir by the Italian Gabriele de’ Mussi. This memoir has been published several times in its original Latin (2,3) and has Historical Background to the Siege of Caffa recently been translated into English (4) (although brief pas- Caffa (now Feodosija, Ukraine) was established by Genoa sages have been previously published in translation, see refer- in 1266 by agreement with the Kahn of the Golden Horde (15). ence [5]). This narrative contains some startling assertions: It was the main port for the great Genoese merchant ships (16– that the Mongol army hurled plague-infected cadavers into the 20), which connected there to a coastal shipping industry to besieged Crimean city of Caffa, thereby transmitting the dis- Tana (now Azov, Russia) on the Don River. Trade along the ease to the inhabitants; and that fleeing survivors of the siege Don connected Tana to Central Russia, and overland caravan spread plague from Caffa to the Mediterranean Basin. If this routes linked it to Sarai and thence to the Far East (12,19,20). account is correct, Caffa should be recognized as the site of the Relations between Italian traders and their Mongol hosts most spectacular incident of biological warfare ever, with the were uneasy, and in 1307 Toqtai, Kahn of the Golden Horde, Black Death as its disastrous consequence. After analyzing arrested the Italian residents of Sarai, and besieged Caffa. The these claims, I have concluded that it is plausible that the bio- logical attack took place as described and was responsible for infecting the inhabitants of Caffa; however, the event was unimportant in the spread of the plague pandemic.

Origin of the 14th-Century Pandemic The disease that caused this catastrophic pandemic has, since Hecker (6), generally been considered to have been plague, a zoonotic disease caused by the gram-negative bacte- rium Yersinia pestis, the principal reservoir for which is wild rodents (7–11). The ultimate origin of the Black Death is uncertain—China, Mongolia, India, central Asia, and southern Russia have all been suggested (see Norris [1] for a discussion of the various theories). Known 14th-century sources are of

Figure 1. Tentative chronology of the initial spread of plague in the mid- *University of California, Davis, California USA 14th century (12–14).

Emerging Infectious Diseases • Vol. 8, No. 9, September 2002 971 HISTORICAL REVIEW cause was apparently Toqtai’s displeasure at the Italian trade in Turkic slaves (sold for soldiers to the Mameluke Sultanate). The Genoese resisted for a year, but in 1308 set fire to their city and abandoned it. Relations between the Italians and the Golden Horde remained tense until Toqtai’s death in 1312 (19). Toqtai’s successor, Özbeg, welcomed the Genoese back, and also ceded land at Tana to the Italians for the expansion of their trading enterprise. By the 1340s, Caffa was again a thriv- ing city, heavily fortified within two concentric walls. The inner wall enclosed 6,000 houses, the outer 11,000. The city’s population was highly cosmopolitan, including Genoese, Venetian, Greeks, Armenians, Jews, Mongols, and Turkic peo- ples (21) In 1343 the Mongols under Janibeg (who succeeded Özbeg in 1340) besieged Caffa and the Italian enclave at Tana (12), following a brawl between Italians and Muslims in Tana. The Italian merchants in Tana fled to Caffa (which, by virtue of its location directly on the coast, maintained maritime access despite the siege). The siege of Caffa lasted until February 1344, when it was lifted after an Italian relief force killed 15,000 Mongol troops and destroyed their siege machines (21). Janibeg renewed the siege in 1345 but was again forced to lift it after a year, this time by an epidemic of plague that devastated his forces. The Italians blockaded Mongol ports, forcing Janibeg to negotiate, and in 1347 the Italians were allowed to reestablish their colony in Tana (19).

Gabriele de’ Mussi Gabriele de’ Mussi, born circa 1280, practiced as a notary in the town of Piacenza, over the mountains just north of Figure 2. The first page of the narrative of Gabriele de’ Mussi. At the top Genoa. Tononi summarizes the little we know of him (3). His of the page are the last few lines of the preceding narrative; de’ Mussi’s practice was active in the years 1300–1349. He is thought to begins in the middle of the page. The first three lines, and the large “A” are in red ink, as are two other letters and miscellaneous pen-strokes; have died in approximately 1356. otherwise it is in black ink. Manuscript R 262, fos 74r; reproduced with Although Henschel (2) thought de’ Mussi was present at the permission of the Library of the University of Wroclaw, Poland. the siege of Caffa, Tononi asserts that the Piacenza archives contain deeds signed by de’ Mussi spanning the period 1344 The narrative begins with an apocalyptic speech by God, through the first half of 1346. While this does not rule out lamenting the depravity into which humanity has fallen and travel to Caffa in late 1346, textual evidence suggests that he describing the retribution intended. It goes on: did not. He does not claim to have witnessed any of the Asian “…In 1346, in the countries of the East, countless numbers events he describes and often uses a passive voice for descrip- of Tartars and Saracens were struck down by a mysterious ill- tions. After describing the siege of Caffa, de’Mussi goes on to ness which brought sudden death. Within these countries say, “Now it is time that we passed from east to west to discuss broad regions, far-spreading provinces, magnificent kingdoms, all the things which we ourselves have seen…” cities, towns and settlements, ground down by illness and devoured by dreadful death, were soon stripped of their inhab- The Narrative of Gabriele De’ Mussi itants. An eastern settlement under the rule of the Tartars The de’ Mussi account is presumed to have been written in called Tana, which lay to the north of Constantinople and was 1348 or early 1349 because of its immediacy and the narrow much frequented by Italian merchants, was totally abandoned time period described. The original is lost, but a copy is after an incident there which led to its being besieged and included in a compilation of historical and geographic attacked by hordes of Tartars who gathered in a short space of accounts by various authors, dating from approximately 1367 time. The Christian merchants, who had been driven out by (Figure 2). The account begins with an introductory comment force, were so terrified of the power of the Tartars that, to save by the scribe who copied the documents: “In the name of God, themselves and their belongings, they fled in an armed ship to Amen. Here begins an account of the disease or mortality Caffa, a settlement in the same part of the world which had which occurred in 1348, put together by Gabrielem de Mussis been founded long ago by the Genoese. of Piacenza.”

972 Emerging Infectious Diseases • Vol. 8, No. 9, September 2002 HISTORICAL REVIEW

“Oh God! See how the heathen Tartar races, pouring lence, and their inhabitants, both men and women, died together from all sides, suddenly invested the city of Caffa and suddenly. And when one person had contracted the illness, he besieged the trapped Christians there for almost three years. poisoned his whole family even as he fell and died, so that There, hemmed in by an immense army, they could hardly those preparing to bury his body were seized by death in the draw breath, although food could be shipped in, which offered same way. Thus death entered through the windows, and as them some hope. But behold, the whole army was affected by cities and towns were depopulated their inhabitants mourned a disease which overran the Tartars and killed thousands upon their dead neighbours.” (Reproduced with permission from thousands every day. It was as though arrows were raining Horrox, pp. 16–20 [4]) down from heaven to strike and crush the Tartars’ arrogance. The account closes with an extended description of the All medical advice and attention was useless; the Tartars died plague in Piacenza, and a reprise of the apocalyptic vision with as soon as the signs of disease appeared on their bodies: swell- which it begins. ings in the armpit or groin caused by coagulating humours, followed by a putrid fever. Commentary “The dying Tartars, stunned and stupefied by the immen- In this narrative, de’ Mussi makes two important claims sity of the disaster brought about by the disease, and realizing about the siege of Caffa and the Black Death: that plague was that they had no hope of escape, lost interest in the siege. But transmitted to Europeans by the hurling of diseased cadavers they ordered corpses to be placed in catapults1 and lobbed into into the besieged city of Caffa and that Italians fleeing from the city in the hope that the intolerable stench would kill Caffa brought it to the Mediterranean ports. everyone inside.2 What seemed like mountains of dead were thrown into the city, and the Christians could not hide or flee Biological Warfare at Caffa or escape from them, although they dumped as many of the de’ Mussi’s account is probably secondhand and is uncor- bodies as they could in the sea. And soon the rotting corpses roborated; however, he seems, in general, to be a reliable tainted the air and poisoned the water supply, and the stench source, and as a Piacenzian he would have had access to eye- was so overwhelming that hardly one in several thousand was witnesses of the siege. Several considerations incline me to in a position to flee the remains of the Tartar army. Moreover trust his account: this was probably not the only, nor the first, one infected man could carry the poison to others, and infect instance of apparent attempts to transmit disease by hurling people and places with the disease by look alone. No one biological material into besieged cities; it was within the tech- knew, or could discover, a means of defense. nical capabilities of besieging armies of the time; and it is con- “Thus almost everyone who had been in the East, or in the sistent with medieval notions of disease causality (22). regions to the south and north, fell victim to sudden death after Tentatively accepting that the attack took place as contracting this pestilential disease, as if struck by a lethal described, we can consider two principal hypotheses for the arrow which raised a tumor on their bodies. The scale of the entry of plague into the city: it might, as de’ Mussi asserts, mortality and the form which it took persuaded those who have been transmitted by the hurling of plague cadavers; or it lived, weeping and lamenting, through the bitter events of might have entered by rodent-to-rodent transmission from the 1346 to 1348—the Chinese, Indians, Persians, Medes, Kurds, Mongol encampments into the city. Armenians, Cilicians, Georgians, Mesopotamians, Nubians, Diseased cadavers hurled into the city could easily have Ethiopians, Turks, Egyptians, Arabs, Saracens and Greeks (for transmitted plague, as defenders handled the cadavers during almost all the East has been affected)—that the last judgement disposal. Contact with infected material is a known mecha- had come. nism of transmission (8–11); for instance, among 284 cases of “…As it happened, among those who escaped from Caffa plague in the United States in 1970–1995 for which a mecha- by boat were a few sailors who had been infected with the poi- nism of transmission could be reasonably inferred, 20% were sonous disease. Some boats were bound for Genoa, others thought to be by direct contact (24). Such transmission would went to Venice and to other Christian areas. When the sailors have been especially likely at Caffa, where cadavers would reached these places and mixed with the people there, it was as have been badly mangled by being hurled, and many of the if they had brought evil spirits with them: every city, every set- defenders probably had cut or abraded hands from coping with tlement, every place was poisoned by the contagious pesti- the bombardment. Very large numbers of cadavers were possi- bly involved, greatly increasing the opportunity for disease 1Technically trebuchets, not catapults. Catapults hurl objects by the transmission. Since disposal of the bodies of victims in a release of tension on twisted cordage; they are not capable of hurling major outbreak of lethal disease is always a problem, the Mon- loads over a few dozen kilograms. Trebuchets are counter-weight- driven hurling machines, very effective for throwing ammunition weigh- gol forces may have used their hurling machines as a solution ing a hundred kilos or more (22). to their mortuary problem, in which case many thousands of cadavers could have been involved. de’ Mussi’s description of 2Medieval society lacked a coherent theory of disease causation. Three notions coexisted in a somewhat contradictory mixture: 1) disease was “mountains of dead” might have been quite literally true. a divine punishment for individual or collective transgression; 2) dis- Thus it seems plausible that the events recounted by de’ ease was the result of "miasma," or the stench of decay; and 3) disease Mussi could have been an effective means of transmission of was the result of person-to-person contagion (23).

Emerging Infectious Diseases • Vol. 8, No. 9, September 2002 973 HISTORICAL REVIEW plague into the city. The alternative, rodent-to-rodent transmis- Caffa, and its gruesome finale, thus are unlikely to have been sion from the Mongol encampments into the city, is less likely. seriously implicated in the transmission of plague from the Besieging forces must have camped at least a kilometer away Black Sea to Europe. from the city walls. This distance is necessary to have a healthy margin of safety from arrows and artillery and to pro- Conclusion vide space for logistical support and other military activities Gabriele de’ Mussi’s account of the origin and spread of between the encampments and the front lines. Front-line loca- plague appears to be consistent with most known facts, tion must have been approximately 250–300 m from the walls; although mistaken in its claim that plague arrived in Italy trebuchets are known from modern reconstruction to be capa- directly from the Crimea. His account of biological attack is ble of hurling 100 kg more than 200 m (25), and historical plausible, consistent with the technology of the time, and it sources claim 300 m as the working range of large machines provides the best explanation of disease transmission into (26). Thus, the bulk of rodent nests associated with the besieg- besieged Caffa. This thus appears to be one of the first biolog- ing armies would have been located a kilometer or more away ical attacks recorded (22) and among the most successful of all from the cities, and none would have likely been closer than time. 250 m. Rats are quite sedentary and rarely venture more than a However, it is unlikely that the attack had a decisive role in few tens of meters from their nest (27,28). It is thus unlikely the spread of plague to Europe. Much maritime commerce that there was any contact between the rat populations within probably continued throughout this period, from other and outside the walls. Crimean ports. Overland caravan routes to the Middle East Given the many uncertainties, any conclusion must remain were also unaffected. Thus, refugees from Caffa would most tentative. However, the considerations above suggest that the likely have constituted only one of several streams of infected hurling of plague cadavers might well have occurred as de’ ships and caravans leaving the region. The siege of Caffa, for Mussi claimed, and if so, that this biological attack was proba- all of its dramatic appeal, probably had no more than anecdotal bly responsible for the transmission of the disease from the importance in the spread of plague, a macabre incident in terri- besiegers to the besieged. Thus, this early act of biological fying times. warfare, if such it were, appears to have been spectacularly Despite its historical unimportance, the siege of Caffa is a successful in producing casualties, although of no strategic powerful reminder of the horrific consequences when disease importance (the city remained in Italian hands, and the Mon- is successfully used as a weapon. The Japanese use of plague gols abandoned the siege). as a weapon in World War II (29) and the huge Soviet stock- piles of Y. pestis prepared for use in an all-out war (30) further Crimea as the Source of European remind us that plague remains a very real problem for modern and Near Eastern Plague arms control, six and a half centuries later (31). There has never been any doubt that plague entered the Mediterranean from the Crimea, following established mari- Acknowledgments time trade routes. Rat infestations in the holds of cargo ships Thanks to Christina Frei for translation from German (Henschel) would have been highly susceptible to the rapid spread of and to Remo Morelli for translation from Italian (Tononi). plague, and even if most rats died during the voyage, they This research was supported by a grant from the University of would have left abundant hungry fleas that would infect California Institute on Global Conflict and Cooperation. humans unpacking the holds. Shore rats foraging on board recently arrived ships would also become infected, transmit- Dr. Wheelis is Senior Lecturer in Microbiology at the University ting plague to city rat populations. of California, Davis. He was trained as a bacterial physiologist and Plague appears to have been spread in a stepwise fashion, geneticist, but for more than 10 years, his research has focused on the on many ships rather than on a few (Figure 1), taking over a history and control of biological weapons. year to reach Europe from the Crimea. This conclusion seems fairly firm, as the dates for the arrival of plague in Constanti- References nople and more westerly cities are reasonably certain. Thus 1. Norris J. East or West? The geographic origin of the Black Death. Bull de’ Mussi was probably mistaken in attributing the Black Hist Med 1977; 51:1–24. Death to fleeing survivors of Caffa, who should not have 2. Henschel AW. Document zur Geschichte des schwarzen Todes. Archives für die gesammte Medizin 1842;2:26–59. needed more than a few months to return to Italy (16). 3. Tononi AG. La Peste Dell’ Anno 1348. Giornale Ligustico de Archeolo- Furthermore, a number of other Crimean ports were under gia, Storia e Letteratura 1884;11:139–52. Mongol control, making it unlikely that Caffa was the only 4. Horrox R, editor. The Black Death. Manchester: Manchester University source of infected ships heading west. And the overland cara- Press; 1994. p. 14–26. van routes to the Middle East from Serai and Astrakhan 5.Derbes VJ. de Mussis and the great plague of 1348. JAMA 1966;196:179–82. insured that plague was also spreading south (Figure 1), 6. Hecker JFC. The epidemics of the Middle Ages. London: Sydenham whence it would have entered Europe in any case. The siege of Society; 1844.

974 Emerging Infectious Diseases • Vol. 8, No. 9, September 2002 HISTORICAL REVIEW

7. Pollitzer R. Plague. Geneva: World Health Organization; 1954. 23. Slack P. Responses to plague in early modern Europe: the implications of 8. Benenson AS. Control of communicable diseases manual. Washington: public health. In: Mack A, editor. Time of plague; the history and social American Public Health Association; 1995. consequences of lethal epidemic disease. New York: New York Univer- 9. Bottone EJ. Francisella tularensis, Pasteurella, and Yersinia pestis. In: sity Press; 1991. p. 111–31. Gorbach SL, Bartlett JG, Blacklow NR, editors. Infectious diseases. Phil- 24. Centers for Disease Control and Prevention. Prevention of plague: recom- adelphia: WB Saunders; 1998: p. 1819–24. mendations of the Advisory Committee on Immunization Practices 10. Dennis DT, Gratz N, Poland JD, Tikhomirov E. Plague Manual: Epidemi- (ACIP). MMWR Morb Mortal Wkly Rep 1996:45:1–15. ology, distribution, surveillance and control. Geneva: World Health Orga- 25. Hadingham, E. Ready, aim, fire! A risky experiment reveals how medi- nization; 1999. eval engines of war brought down castle walls. Smithsonian 1975;30:78– 11. Boyce JM. Yersinia species. In: Mandell GL, Douglas RG Jr, Bennett JE, 87. editors. Principles and practice of infectious disease. 2 ed. New York: 26. Payne-Gallwey R. A summary of the history, construction and effects in John Wiley and Sons; 1985. p.1296–301. warfare of the projectile-throwing engines of the ancients, with a treatise 12. Dols MW. The Black Death in the Middle East. Princeton (NJ): Princeton on the structure, power and management of Turkish and other oriental University Press; 1977. bows of medieval and later times. London: Longmans, Green and Co.; 13. Gasquet FA. The great pestilence (A. D. 1348–9), now commonly known 1907. as the Black Death. London: Simpkin Marshall, Hamilton, Kent & Co.; 27. Twigg G. The Black Death: a biological reappraisal. London: Batsford 1893. Academic and Educational; 1984. 14. Bartsocas CS. Two fourteenth century Greek descriptions of the "Black 28. Barnett SA. The rat: a study in behavior. Chicago: University of Chicago; Death.” J Hist Med Allied Sc 1966; 21:394–400. 2000. 15. Vasiliev AA. The Goths in the Crimea. Cambridge (MA): Mediaeval 29. Harris SH. Factories of death: Japanese biological warfare 1932–45 and Academy of America; 1936. the American cover-up. New York: Routledge; 1994. 16. Gardiner R. The Age of the galley: Mediterranean oared vessels since 30. Alibek K, Handelman S. Biohazard: The chilling true story of the largest pre-classical times. Annapolis (MD): Naval Institute Press; 1995. covert biological weapons program in the world—told from the inside by 17. Fayle CE. A short history of the world's shipping industry. London: the man who ran it. New York: Random House; 1999. George Allen & Unwin; 1933. 31. Inglesby TV, Dennis DT, Henderson DA, Bartlett JG, Ascher MS, Eitzen 18. Lewis AR, Runyan TJ. European naval and maritime history, 300–1500. E, et al. Plague as a biological weapon: medical and public health man- Bloomington (IN): Indiana University Press; 1985. agement. JAMA 2000;283:2281–90. 19. Grousset R. The empire of the steppes: a history of Central Asia. New Brunswick (NJ): Rutgers University Press; 1970. 20. Obolensky D. The Byzantine commonwealth: Eastern Europe, 500–1453. Address for correspondence: Mark Wheelis, Division of Biological Sciences, London: Weidenfeld and Nicolson; 1971. Section of Microbiology, University of California, Davis, One Shields 21. Howorth HH. History of the Mongols, from the 9th to the 19th century. Avenue, Davis, CA 95616-8665 USA; fax: 530-752-363; e-mail: New York: Burt Franklin; 1880. [email protected] 22. Wheelis M. Biological warfare before 1914. In: Geissler E, Moon JEvC, editors. Biological and toxin weapons: research, development and use from the Middle Ages to1945. London: Oxford University Press; 1999. p. 8–34.

Full text free online at www.cdc.gov/eid

The print journal is available at no charge to public health professionals

YES, I would like to receive Emerging Infectious Diseases.

Please print your name and business address in the box and return by fax to 404-371-5449 or mail to EID Editor CDC/NCID/MS D61 1600 Clifton Road, NE Atlanta, GA 30333

Moving? Please give us your new address (in the box) and print the number of your old mailing label here______

Emerging Infectious Diseases • Vol. 8, No. 9, September 2002 975